Pedja said:From your list only DH-200 qualifies like SS power amp. If you had something to report about the sonic differences between unregulated and regulated supplies used with it, maybe with some slight changes in both versions and get some conclusions, you should do that... 🙄 Then I or someone else might be very glad to jump on your neck shouting some stupidity like "it is ridicilous to say it is original"(I mean stupidity beacuse you have not said that...
) and so on and so on...
Then after all your trials and errors and successes you had
, you could make similar report about the buffer, that you've used with power IC back in the 70's altogether with regulated supply, just like a lot of other people did then, so we could go another round of jumping on the neck...😎
Pedja
You do miss the point here. I'm not interested it recalling subjective impressions about old amps from the past. I'm just saying it's been done before, with all kinds of circuits. There is nothing new here.
The point of using a regulator with ANY amp is to provide the amp with clean, stable, quit power. Now the amp can perform at it's best with out concerns about AC line noise or ripple. This mean a better signal to noise ratio and lower noise floor for that "subtle detail". The voltage is stable so the amp will be more stable. That translates into being able to drive difficult loads with less problems. The impedance of the power supply is low across the band width of the amp and this improves the frequancy response of the amp.
As for a buffered input to an inverted chip amp. That gives the input circuit of the chip a stable/predictable impedance. You have control of the impedances around the feedback network. Simple stuff here.
I would suggest you read slowly and carefully and then think before making statments that may make you look bad.
Later BZ
nearer to Blackgate
Dear all
I wanna add in a regulator but which is nearer to Blackgate ? (Since I can’t find those spec)
For 35v Ruby ZL the ripple current is 2770 ma while the impedance is 0.021ohm and for Pana FC 35v the ripple current is 1945 ma while the impedance is 0.03 ohm.
Both have the same capacitance which is 1000uF.
Base on data it looks like Ruby is a clearer winner, but, I have not use them. I only managed to get Pana FC.
Please advice.
Warmest regards
Dear all
I wanna add in a regulator but which is nearer to Blackgate ? (Since I can’t find those spec)
For 35v Ruby ZL the ripple current is 2770 ma while the impedance is 0.021ohm and for Pana FC 35v the ripple current is 1945 ma while the impedance is 0.03 ohm.
Both have the same capacitance which is 1000uF.
Base on data it looks like Ruby is a clearer winner, but, I have not use them. I only managed to get Pana FC.
Please advice.
Warmest regards
For a regulated supply, you will need either some 4700 uFs of 10,000 uF's which are placed before the regulator circuit. Then you use something like 33 uF-100uF caps on the pins of the chip.
The actual circuit is on
Pedja's site
Panasonic FC's are good capacitors but not available in the higher values so you will need to find somethng that is available to you locally. In the UK BHC Aerovox types are quite popular.
If you have not made a Gainclone before, please build one without the regulators first (use 1000 uF Panasonic FCs on the pins). Get that working properly and then add the regulator circuit. You will only need to remove the 1000 uFs and replace with the 33-100 uFs. 😉
Just want to share my experience with regulated GC. I orginally have a regular NIGC (3875) driving a pair of 4 ohms speaker, the bass is deep but not tight even at low volume. I then added 10000uF at PS. The bass improved a bit, but still not tight enough, also the mid and highs were not good anymore.
Then I added LT1083 at the PS and changed the 3875's capacitors to 100uF. Now I am very satisfy with the result : bass is tight, mids and highs are also very good.
You can check out my schematic and other details in my web page.
I measured by a multimeter that there is a sag in supply voltage (up to 10% drop) if I have no regulator even I add lots of caps.
The voltage stay very stable after I added regulator. I am guessing that the voltage drop at PS will affect the wattage available to the chip and thus affecting the sound.
I have not tried adding buffer yet because I am so busy listening to my amp now!
BTW, in what aspect would a buffer affect the sound?
Alex
Then I added LT1083 at the PS and changed the 3875's capacitors to 100uF. Now I am very satisfy with the result : bass is tight, mids and highs are also very good.
You can check out my schematic and other details in my web page.
I measured by a multimeter that there is a sag in supply voltage (up to 10% drop) if I have no regulator even I add lots of caps.
The voltage stay very stable after I added regulator. I am guessing that the voltage drop at PS will affect the wattage available to the chip and thus affecting the sound.
I have not tried adding buffer yet because I am so busy listening to my amp now!
BTW, in what aspect would a buffer affect the sound?
Alex
HDTVman said:
I'm gonna repead this for the 1000th time:
I'M NOT DOING ANYTHING ORIGINAL
Get it?
What's the problem of reporting improvements with regulated PSU and these chips?
I'm tired...
Indeed, King Arthur had a horse called Gainclone which he fed every day.
But that was a regular supply not a regulated supply! 😀
The Sufis tell us that all disharmony comes from the illusion of self! This thread is a good example of that philosophy.
Kahlil Gibran said of children, "they come through you but not of you". The same is true of ideas! 🙂
But that was a regular supply not a regulated supply! 😀
The Sufis tell us that all disharmony comes from the illusion of self! This thread is a good example of that philosophy.
Kahlil Gibran said of children, "they come through you but not of you". The same is true of ideas! 🙂
HDTVman said:
Yes, simple stuff here.😎
And you describe it quite well.😉
I wonder why it's so hard for some guys to understand...
Some people who argue that this should not be better blah blah blah should at least read your post.
Thanks.
alexw88 said:
Alex, your impressions are exactly what I described when I listened to my regulated PSU.
I have the same feelings.
This thing seams to drive any speaker (better still the LM3886, double the current).
Nice work.😎
Variac, as your post went to Texas, I just want to clarify some things:
"Greg knows exactly what he is talking about. You are attacking someone just because your ego is so sensitive?"
I'm not attacking, I'm defending myself.
I don't know where he came from, or who sent him.
ONLY the moderators can warn me, if they need to.
Clear?
"Peter has to just immedietely accept that your amp is the "real" high end?"
Of course not.
But he can't say it's not better without LISTENING.
I know all that PSRR talk with my eyes closed but in this case it's got nothing to do with it.
You guys should get employed on audio magazines and review some amps without listening.
"Greg knows exactly what he is talking about. You are attacking someone just because your ego is so sensitive?"
I'm not attacking, I'm defending myself.
I don't know where he came from, or who sent him.
ONLY the moderators can warn me, if they need to.
Clear?
"Peter has to just immedietely accept that your amp is the "real" high end?"
Of course not.
But he can't say it's not better without LISTENING.
I know all that PSRR talk with my eyes closed but in this case it's got nothing to do with it.
You guys should get employed on audio magazines and review some amps without listening.
I hadn't thought of this before but how many of you who have built regulated Gainclones have used just one regulator circuit (like Alex) and how many have two?
I have built two regulator circuits, the idea being to position each as closely to the pins of each chip as I can. I also have the option of using a second 300 VA transformer/rectifier if I need it but am using one to begin with.
If one regulator circuit will suffice, then it obviously halves the cost of a regulated supply. And looking at how Alex has laid out his amp, placing the regulator cicuit between the chips still keeps the connection reasonably short.
I have built two regulator circuits, the idea being to position each as closely to the pins of each chip as I can. I also have the option of using a second 300 VA transformer/rectifier if I need it but am using one to begin with.
If one regulator circuit will suffice, then it obviously halves the cost of a regulated supply. And looking at how Alex has laid out his amp, placing the regulator cicuit between the chips still keeps the connection reasonably short.
Nuuk said:
Nuuk, as you can see on my pics, I use only one pair for both channels, no problems.
With a small cap (I used 47nf) at the reg's output, no problem to have a little more wire.
Anyway, I have between 10 and 15cm of wire between the regs and the chips' PCB.
No harm to double the regs, but when I saw the pics of your boards I thought you were going to make a pair of monoblocks...😀
No, both channels will be in one case.
Of course, another factor to remember (as Pedja has rightly pointed out) is that using two circuits halves the current drawn through each regulator.
So, if a single regulator circuit is used, it may be better to take Neal's advice and go for the LM338K instead of the LM338T. Although more expensive, the cost is offset by not needing four of them (and fewer caps).
(Still awaiting delivery from Farnell)
Of course, another factor to remember (as Pedja has rightly pointed out) is that using two circuits halves the current drawn through each regulator.
So, if a single regulator circuit is used, it may be better to take Neal's advice and go for the LM338K instead of the LM338T. Although more expensive, the cost is offset by not needing four of them (and fewer caps).
(Still awaiting delivery from Farnell)
I'm posting a start of a board layout for the gainclone or whatever we are calling it now.
It is very incomplete, but I would like to get opinions before I go much further. I have never done a circuit board b4, so be kind. I may very well have made a really colossal newbie error or two.
Since I don't have any board layout software, I have done it rather tediously in Corel Draw. I have looked up various componets on catalogs to get the dimensions.
At the moment the paths are color coded so I can better keep track of which trace is which.
It is very incomplete, but I would like to get opinions before I go much further. I have never done a circuit board b4, so be kind. I may very well have made a really colossal newbie error or two.
Since I don't have any board layout software, I have done it rather tediously in Corel Draw. I have looked up various componets on catalogs to get the dimensions.
At the moment the paths are color coded so I can better keep track of which trace is which.
An externally hosted image should be here but it was not working when we last tested it.
Nuuk said:
Nuuk, independent regs per channel is better, of course, but single regs work fine for both channels.
The LM338 has a max current output of 12A, very good indeed.
As this is not a class-A amp, an music is not a continuous tone, this is enought for both channels.
Much better than unregulated, if you ask me.😀
Everybody is free, of course, to use independent diodes/caps/regs and even trafo per channel.😎
That would not me "mini" anymore, but more like a Krell.
geewhizbang said:
gainregbufclone.😀
Your layout is fine, but you don't need 5w resistors😱 .
100 ohm 0.5 or 1 watt resistors are fine there.
The output of the op-amp goes where?😕
No trace?
How is this thread useful to anyone hoping to do any DIY?? It needs a damn good edit to get it back into any sort of useful shape (you can edit this post out too while you're at it).
I agree with several posts that point out that much of DIYaudio.com is not really about DIY audio anymore...
I agree with several posts that point out that much of DIYaudio.com is not really about DIY audio anymore...
Variac said:
Of course, if you stray a few kilometers from Lasceaux you will wind up at Les Eyzies de Tayac, and if you wind up at Les Eyzies you will undoubtedly have dinner at Centennaire, and if you have dinner at Centennaire you will undoubtedly drink too much St. Emillion or Margaux in which case everything will look like a single-chip amplifier or solid state regulator. this becomes even more problematic if you have succombed to the product of the graveuse!
why use a discrete regulator? , particularly at the edge of its performance envelope? why not build and demonstrate a low noise, low dropout, high power, well compensated regulator circuit instead of using a discrete device.
If the benefit to the sound is as great as Carlos suggests, by the same logic wouldn't it be better still to employing the super-regulator techniques here? Even a Sulzer is going to beat the pants off a discrete.
jackinnj said:
I don't say this is the best regulated PSU for this.
But it's so much better than unregulated...
Pedja says discrete is very good too (if not better).
tbla likes the (expen$ive-and-hard-to-find-here) LT1083.
Be creative, make your own, make it better, post it.
It's fun.😎
- Status
- Not open for further replies.